REPORT ON THE ECHINOIUEA. 61 



The Challenger having collected a number of large specimens (PI. V. fig. 1) fully 

 equalling in size the original specimen of Michelin, I am now able to clear up many 

 points left doubtful, and to describe at length this interesting Sea-urchin. 



On the actinal side the spines are of two kinds both in the ambulacral and interam- 

 bulacral areas. First, small, slender, more or less cylindrical, pointed, finely serrated, 

 secondary spines (PL VI. figs 20c and 21c) surmount the secondary tubercles intervening 

 between the primary rows, especially on the edge of the interambulacral plates adjoin- 

 ing the poriferous zone ; secondly, the primary spines which, near the apical system are 

 scarcely larger than the secondary spines, but are readily distinguished by their triangular 

 or flattened shape, though the ornamentation of the smaller primaiy radioles is similar to 

 that of the secondaries. As they become large the base of the shaft becomes more finely 

 granular, and the extremity of the spines fluted, ^\^th serrated edges ; this fluting and 

 serration exists only in the largest primary radioles, and is limited to the part of the 

 shaft included between the milled ring and the shoe. The spines increase very rapidly 

 in size, so that before reaching the ambitus there are in the ambulacral area from four to 

 five, more or less curved, varying in length from nearly the diameter of the test to fully 

 three times as much (PI. V. fig. 1, PI. VI. figs. 19a, b, c, 22a, h, c), and two to three of like 

 size on the ambulacral plates near the ambitus. Abnormal as these radioles seem at 

 first glance, the relationship of the genus to the Arbaciadse gives a very simple expla- 

 nation of the enormous growth of some of the primarj^ spines. The shaft of the larger 

 primary radioles is composed of two very distinct parts, the basal part, the extension of 

 the shaft beyond the collar, which is triangular, passing into an hexagonally fluted or 

 angular shaft (PI. VI. fig. 22a', c'), finely striated and more or less distinctly covered with 

 minute serrations on the edge of the fluting [which in the shorter primary spines extends 

 to the very tip or near its extremity (PL VI. fig. 19c)] surmounted by the smooth 

 polished portion of the shaft (PI. VI. figs. 19a, 22a, c), which is rarely straight, and 

 often very considerably curved, and which sometimes is twice as long as the diameter 

 of the test, and sometimes forms a mere tip to the coloured angular base (PL VI. figs. 

 19d, 22d). This tip is angular or triangular in younger specimens. As the primary 

 spines pass towards the ambitus they rapidly become shorter, flatter, more slender (PL VI. 

 figs. 19cZ, 22d), straighter, and immediately around the actinal membrane are reduced 

 again to slender spines, not more powerful than the secondary ones (PL VI. figs. 20c, 21c). 



It is the identity of the smaller flattened tipped spines (PL VI. figs. 20&, 21 6) 

 around the actinostome with those of other Arbaciadse which shows plainly that the 

 principal part of the shaft of the radioles of this genus consists of an exaggerated 

 growth of the cap at the tip of the spine so characteristic of the Arbaciadse and 

 to which Desmoulins (Actes Soc. Lin. de Bordeaux, 1870) first called attention. This 

 cap, however, in the other species of the family is found only on the spines of the 

 actinal surface, and I inferred from the peculiar mode of locomotion of the common 



